There’s a very good op-ed about why universities charge tuition to Ph.D. students at all (boldface mine):

It’s not the norm for PhD students to pay any tuition. Where I’m in graduate school, at Yale, the billing and payment for my tuition happen completely out of my view. The money moves from one institutional account to another. I’m never responsible for it, and I never see any of the money that I’ll supposedly be taxed on. I am currently in the third year of my PhD studies, and while I do teach, I am not taking any classes. It is unlikely that I will take any more classes while at Yale. Yet I’m still theoretically being “charged” tuition on some university administrator’s spreadsheet — and the GOP legislation would treat the waiver of that tuition as a taxable benefit.

In fact, though, not making graduate students pay tuition reflects the material reality of graduate school: We do valuable research and teaching labor for the university. It would be preposterous to bill us for this — akin to asking medical residents to pay the hospital for the right to train there. The fact that the university doesn’t actually make most of us pay the tuition it ostensibly charges is a tacit admission of this fact…

Like grants, tuition is another way the university makes money from science. While tuition looks mainly like an artifact of accounting for most graduate students in the humanities and social sciences (where the university just pays itself for tuition), in the natural and applied sciences, tuition is a way for the university administration to get more money out of the public. As the university bulletin explains, “for a standard [research assistant] appointment in addition to the salary, the grant pays half of the tuition.” In other words, the government — through the National Science Foundation, NIH, NASA and other agencies — is transferring $20,500 per year per research assistant into Yale’s accounts, at a time when the university endowment is at an all-time high above $27 billion. This tuition transfer certainly totals in the millions every year for Yale alone, and must add up to a mind-boggling sum across the country. Again, this is money that the public pays universities for granting graduate students the privilege of working for them to create more wealth for those same universities.

Graduate students are labor–to deny otherwise is ridiculous. Tuition is another way, in the sciences, to hit funding agencies up for more money, while simultaneously being able to claim that graduate students are students, and not providing important labor (akin to that of ‘student-athletes’ at high-powered sports schools). That said, it goes off the rails when discussing overheads.

Before I get to how it goes off the rails, it’s important to understand why this seemingly arcane bit of accounting matters: many universities are not negotiating in good faith with their students, so they will use inaccuracies to obscure the debate and undermine the legitimacy of the students (“Those kids don’t understand how university budgets work!”). So getting this right matters, and this is not correct:

Science labs on university campuses like where I work are actually sources of income for universities. For most of what goes on in a university science department, the public is footing the bill. Professors apply for research grants to outside foundations and government agencies, most significantly the National Science Foundation and the National Institutes of Health. In the grant application, the professor proposes the budget for the laboratory, including equipment, supplies, and salary and benefits for whoever works in the lab — including graduate student researchers like me. Universities then take a huge amount off the top. Yale, for example, charges the public an additional 67.5 percent “overhead” fee on grants that faculty bring in. So if the National Science Foundation approves a proposal for a $100,000 project, the agency writes a check for $167,500, with that extra margin going to the university for “facility and administrative costs.”

Nope, nope, nope. As we’ve discussed before, about 28-30% of total grant dollars go to what are referred to as “total indirect costs”, which means for every $100,000, the university receives an additional $35,000-$40,000. That money goes to everything from the evil central administration (which, I’ll grant at too many places is bloated) to paying for payroll taxes, to the depreciation of the physical plant (laboratories, etc.), along with a bunch of other stuff that is related but can’t be put into grants as line items. Thing is, this rate is essentially constant across all universities. So something weird is going on: my guess is some places are losing money and others are making out like bandits. On the other hand, we shouldn’t expect government contractors–that’s what federally funded researchers are–to take a loss either.

So, yes, graduate student tuition puts the lie to notion that students aren’t labor. But don’t muck up a good argument with errors about overheads.